4. EXPLORING THE DYNAMICS OF HEALTH AND WEALTH USING LINKED ADMINISTRATIVE DATA This project focuses on the causal links between socioeconomic status (SES) - in particular, financial and housing wealth - and health status. A first goal is to create a long panel dataset with detailed information on health and SES at the individual and household levels, using all existing waves of the Health and Retirement Study (HRS), thus covering the U.S. population aged 50 and older. A key innovation is the use of administrative, and therefore highly reliable, data on medical conditions and expenditure that are newly available from Medicare Part A/B claims data merged to the HRS sample at the individual level. We will then develop an econometric model of health dynamics that establishes hazard rates for the onset of health conditions and identifies patterns of interdependence between onsets, and that can be used as a predictive model of future prevalence of health conditions. This model will be estimated using a detailed dataset on health conditions constructed from a separate, large panel dataset based on Medicare A/B/D claims and auxiliary information from the National Health and Nutrition Examination Survey (NHANES). This research will expand existing models of the joint health-wealth dynamics, adding the newly developed econometric modeling strategies for the dynamics of health in particular. We will use exogenous variations in wealth to provide additional sources of identification of the direct causal effects of wealth changes on health. Such exogenous variation is provided by differential exposure to wealth shocks in the presence of health care delivery systems that vary in the financial impact of co-payments, particularly for chronic conditions and preventative and palliative therapies. The results will have direct implications for health policy regarding access to medical services, priorities for care for chronic conditions, and behavioral interventions. The project will also deliver detailed models of health dynamics that can be used to simulate life-cycle health under alternative insurance and intervention treatments in order to assess the cost of government health policies and their effects on older Americans'well-being.